Focal interstitial irradiation is being used in the clinical management of malignant brain tumors and generally follows a conventional teletherapy protocol. This approach is effective against certain types of tumors, but has the potential for causing substantial damage to normal brain tissues. An alternative approach is to use focal high-dose external-beam irradiation, i.e., radiosurgery, instead of high activity radioisotopes. However, the deleterious effects of radiosurgery have not been well studied, and critical biological information is lacking, particularly regarding dose-volume relationships, and the extent to which focal radiation-induced injury can be modified. This proposal focuses on three hypotheses related to the response of normal brain tissue to high-dose focal irradiation: 1) the severity of radiation-induced damage is related to the volume of irradiated tissue; 2) subtle changes in morphology/physiology can predict the ultimate tissue breakdown caused by radiosurgery; and, 3) the pathophysiologic consequences of radiosurgery are mediated in part by the presence of polyamines, and reductions in tissue polyamine levels may moderate the extent of radiation-induced injury. The proposed study will involve a dog model and will use well- established noninvasive methodologies to measure volumes of injury, vascular permeability (blood-to-brain, k/i, and brain-to-blood, k/b, transfer constants), regional cerebral blood flow (rCBF), mean transit time (MTT) of blood and vascular volume. Immunohistochemical methods will be used for a qualitative and quantitative appraisal of the cellular response in irradiated brain. The specific aims are: 1) to determine the morphologic and physiologic responses of normal brain as a function of time after a single radiosurgical treatment to a defined volume of normal brain; 2) using multiple quantitative endpoints, to determine the dose response characteristics of normal brain after a single radiosurgical treatment, and quantify how the extent of injury after a specified dose relates to the volume of brain irradiated; 3) in terms of radiation- induced necrosis, to determine if and to what extend dose heterogeneity in the high dose region of the dose distribution affects the dose-volume relationship; 4) to determine if changes in CSF polyamine levels and/or brain tissue density (CT number), blood flow-related parameters (rCBF, MTT, vascular volume), or permeability (k/i and k/b) precede general tissue breakdown and may be useful predictors of the ultimate expression of radiation injury; 5) using immunohistochemical markers specific for macrophages, vascular endothelial cells, astrocytes and cells synthesizing DNA, to qualitatively and quantitatively determine the cellular response of normal brain tissue to focal irradiation; and 6) to determine if and to what extent an intravenous infusion of alpha-difluoromethylornithine (DFMO) modifies the morphologic and physiologic consequences of a standard radiosurgical treatment.